OpenAI Provider
The OpenAI provider implements support for OpenAI's models including GPT-4, GPT-4-turbo, GPT-3.5-turbo, and other variants through the OpenAI API.
Supported Models
python
from aicore.models_metadata import METADATA
# List all supported OpenAI models
openai_models = [model for model in METADATA if model.startswith("openai-")]
print(openai_models)Key Features
- Supports all OpenAI chat and completion models
- Full streaming support with real-time token counting
- Automatic retry handling for rate limits
- Multimodal support (images via base64 encoding)
- Detailed usage tracking and cost calculation
Configuration
python
from aicore.llm.config import LlmConfig
config = LlmConfig(
provider="openai",
api_key="your_api_key", # or set OPENAI_API_KEY environment variable
model="gpt-4o",
temperature=0.7,
max_tokens=4000
)Basic Usage
python
from aicore.llm import Llm
# Initialize with config
llm = Llm(config=config)
# Simple completion
response = llm.complete("Hello world")
print(response)Advanced Usage
Multimodal Input
python
import base64
# Encode image
with open("image.png", "rb") as image_file:
base64_image = base64.b64encode(image_file.read()).decode('utf-8')
response = llm.complete(
[
{"type": "text", "text": "What's in this image?"},
{"type": "image_url", "image_url": f"data:image/png;base64,{base64_image}"}
]
)Error Handling
The provider automatically handles:
- Rate limits (with exponential backoff)
- API key errors
- Model availability issues
- Context window overflows